Abstract Diagnosed in >187,000 persons each year, non?small cell lung carcinoma (NSCLC) has been a leading cause of cancer-related mortality in the US. For NSCLC patients, radiation therapy (RT) or chemoradiotherapy has been used as the standard care when the disease stays at locally advanced or local regional stage. Recently, radiotherapy have also been proven to be a viable alternative to lobectomy and lymph node dissection in stage I NSCLC patients. Despite its wide application in NSCLC patient management, the efficacy of RT would often be limited by the innate or acquired radioresistance. A wide range of radiosensitizers, such as cisplatin, 5-Fu, nicotinamide, and etoposide, are often used in concurrent with RT to improve treatment outcome. Unfortunately, these drugs could also cause severe systematic toxicities while enhancing tumor killing efficacy. This research proposes to develop a new category of radiosensitizer based on the ultra-small metal- intercalated-carbon dots (named as M@Cdots), aiming to achieve enhanced cancer killing effect with minimal systemic toxicity. Our M@Cdots have various unique features including: 1) enhanced RT therapy effects of X- ray: the metal fillings of M@Cdots enhance photoelectric effects of X-ray, which, in conjunction with the carbon surface catalyzed radiolysis, lead to remarkable radiosensitizing effects; 2) limited cytotoxicity: due to the bio- inert carbon shell, M@Cdots are not susceptible to metal falloff as many conventional high-Z nanoparticles are, and they cause little cytotoxicity in the absence of ionizing irradiation. Meanwhile, due to its ultra-small size (3nm), M@Cdots are efficiently excreted through renal clearance with minimal reticuloendothelial system (RES) uptake, reducing the risk of long-term toxicity to the host. 3) template synthesis methods: M@Cdots are made through mesoporous template calcination and are 3 nm in diameter. This unique approach allows easy scale-up synthesis of particles, and permits reliable metal encapsulation without extensively re-exploring synthetic procedures. Our preliminary therapy results are very promising. On this basis, we will also explore active tumor targeting by conjugating neurotensin (NTS) ligands to the surface of M@Cdots. The target, neurotensi receptor 1 (NTSR1), is upregulated in large numbers of lung cancer patients but not in normal lung tissues. It is hypothesized that with excellent tumor selectivity, efficient radiosensitization, minimal metal falloff, and efficient renal clearance, NTS-M@Cdots will lead to greatly improved RT outcomes at the same or even reduced radiation doses while causing minimal systemic toxicities. Although the current study is focused on NSCLC, the methodology can also be easily extended to treatment of other cancer types, for instance head and neck, breast, and prostate cancer.